Development and implementation of novel sensor fusion algorithm for occupancy detection and automation in energy efficient buildings. (January 2019)

• Record Type: Journal Article
• Authors: Roselyn, J. Preetha; Uthra, R. Annie; Raj, Ajith; Devaraj, D.; Bharadwaj, Pranav; Krishna Kaki, Sai Vijay Deeraj
• Is Part Of: Sustainable cities and society. Volume 44(2019)
• Page Start: 85
• Page End: 98
• Publication Date: 2019-01
• Subjects: Occupancy detection -- Sensor fusion -- Image processing -- Energy management -- Distributed computing -- Internet of things; Sustainable urban development -- Periodicals; Sustainable buildings -- Periodicals; Urban ecology (Sociology) -- Periodicals; 307.76
• DOI: 10.1016/j.scs.2018.09.031
• Languages: English
• ISSNs: 2210-6707

Abstract

Highlights: This paper deals with achieving both energy efficiency and comfortness in a particular room by accurate mapping of the environment parameters. This paper proposes novel sensor fusion algorithm for occupancy detection by combining two sensory inputs. The output of camera is processed by background subtraction based image processing algorithm for detecting dynamic events and ROI based segregation algorithm for detecting static events. The two sensory processed inputs are performed in a distributed computing platform in different processors and the outputs of these processing algorithms through IoT are obtained in the master controller. The proposed sensor fusion algorithm helps in increasing output accuracy because of feedback validation. Abstract: This paper deals with achieving both energy efficiency and comfortness in a particular room by controlling lighting and cooling appliances of a building by deploying modern sensors like Thermal sensors and Camera for accurate mapping of the environment parameters. The unique part of this paper is that a Sensor Fusion algorithm has been used in different segments of the room which combine the sensory output from the Thermal Sensor and Camera to ensure that the occupancy detection is done accurately with least amount of fluctuations without any delay. The two algorithms, namely, image processing algorithm and sensor signal processing runs in distributed computing environment in two different processors and the outputs of processing algorithms are received in the master controller through IoT cloud platform. The proposed image processing algorithm is effective in detecting both static and dynamic event changes using background subtraction and ROI based segregation models. The system develops the use of this sensor fusion algorithm to provide an intuitive control of the environment and improve the efficiency of the decision making process using National Instruments MyRIO 1800 and cRIO 9082 processors in LabVIEW platform which also reduces the chances or consequences of any fault occurring in the system.